NERVOUS SYSTEM OF FISHES. 185 



respective differences of structure will be readily appreciated : the 

 crus is a compact tract of medullary with, a small proportion of grey 

 matter ; the nerve is a bundle of nerve filaments : the medullary 

 tract of the crus is fibrous, but the fibres are as fine as in the crus 

 cerebri, and much more numerous and less easily separable than in 

 the true olfactory nerve : in this there is no grey tract ; it consists 

 wholly of comparatively large and readily separable white fibres, 

 which radiate at once upon the olfactory capsule: the divergence 

 and radiation of the true end of the olfactory nerve is well seen in 

 the Lepidosii-en (yjig. 54. l. oZ.). In the Sharks a ventricle is continued 

 to each rhinencephalon along its crus from the prosencephalon. The 

 olfactory nerve never forms a ganglion before spreading upon the 

 olfactory capsule : the rliinencephalic crus, when prolonged to the 

 capsule, always expands into a ' tuberculum olfactorium,' or rhinen- 

 cephalon, before it transmits the true olfactory nerves to the cap- 

 sule. In other words, the olfactory nerve conveys impressions to a 

 proper centre or lobe, which in fishes may be situated close to the 

 capsule, or close to the rest of the brain, and the length of its crus 

 will be inversely as that of the nerve. To say, with Cuvier, that 

 " the ganglion of the olfactory nerve is sometimes at its beginning, 

 and sometimes at its end" (t. iii. xxiii. p. 146.), or that it occurs " in 

 the course of the olfactory nerve, at a greater or less distance from 

 the hemispheres " (xxvii. p. 227.), is, in fact, to deny the marked ana- 

 tomical difference between the crus and the nerve proper ; and to 

 overlook the serial homology of the rhinencephalic crui'a with those 

 of the prosencephalon. The olfactory lobes or rhinencephala them- 

 selves are serially homologous with the optic lobes. As to the pro- 

 sencephalon, since this does not immediately receive or transmit any 

 nerve, it resembles in this important character the cerebellum, and 

 proceeds, even in the present class, to be developed to a degree 

 beyond that of the ganglions of any nerves or organs of sense. 



The more special homology of the prosencephalic lobes, under 

 their normal proportions and solid structure in Osseous Fishes, with 

 the parts of the complex and fully developed prosencephalon in 

 INIammalia, will be made manifest as we trace the progress of that 

 complication synthetically. Cuvier had already, by the opposite 

 course of analysis, reduced tlie hemispheres in birds to the ' corpora 

 striata,' with their commissures and a thin supra-ventricular cover- 

 ing. " Le corps cannelc," he says, " forme a hii seul presque tout 

 Fhemisphere." {Lcgons cTAnat. Comp. t. ii. 1799, p. 162.) But he 

 failed altogether to recognise the homology of the prosencephala in 

 Fishes. Since Arsaki's time their homology with the cerebral lobes 

 of Reptiles, Birds, and ]\lammals has been generally recognised. 



